IP Law Daily, PATENT NEWS: Counsel for biopharma companies spar before Supreme Court over enablement of antibody genus claims, (Mar 27, 2023)

By Thomas Long, J.D.

Justices questioned whether the parties really disagreed on the legal standards, and if not, what the Court could contribute besides dismissing the case as improvidently granted.

The Supreme Court today heard arguments in Amgen Inc. v. Sanofi (No. 21-757), a case that could affect the ways pharmaceutical research companies must develop and describe their inventions in order to receive the broad scope of protection they would prefer to justify their research and development costs. Biopharmaceutical company Amgen Inc. is asking the Court to clarify the test for patent enablement. Amgen maintains that the U.S. Court of Appeals for the Federal Circuit improperly held that two Amgen patent were invalid for lack of enablement because their broad genus claims for antibody compounds for lowering cholesterol were not described in enough detail for people skilled in the relevant art to carry out the full scope of the claimed invention.

Some of the Justices—particularly Justice Gorsuch—seemed to wonder what the legal dispute between the parties was, or whether there even was one. While counsel for both sides said that they generally agreed on the relevant law—including a requirement that an invention is not enabled if the claims require “undue experimentation”—they differed as to how the legal tests should be applied. This appeared to leave some Justices wondering what role the Court could play in the case, and what it should say to lower courts to clarify the law. Counsel for Sanofi and for the federal government both suggested that the Court could dismiss the case as improvidently granted, and the Justices did not seem to reject that suggestion out of hand.

Infringement dispute. A jury determined that respondent Sanofi Aventisub—accused of infringing Amgen’s patent relating to antibodies that help reduce low-density lipoprotein cholesterol (LDL-C), or bad cholesterol—failed to prove that the asserted claims were invalid for lack of written description and enablement. The district court, however, granted Sanofi’s motion for judgment as a matter of law (JMOL) for lack of enablement, after concluding that the patents are invalid. The Federal Circuit affirmed, agreeing that the patent claims are not enabled because they require undue experimentation.

Petition for review. Amgen’s petition for certiorari contained two questions, but the Supreme Court agreed only to consider the second question: Whether enablement is governed by the statutory requirement that the specification teach those skilled in the art to “make and use” the claimed invention, 35 U.S.C. § 112, or whether it must instead enable those skilled in the art “to reach the full scope of claimed embodiments” without undue experimentation—i.e., to cumulatively identify and make all or nearly all embodiments of the invention without substantial ‘time and effort.’”

Section 112. Section 112(a) states that the written description in a patent’s specification must contain enough information to enable a person skilled in the pertinent art “to make and use” the invention.

Wands factors. Courts have added further gloss to Section 112, including a requirement that the written description not require “undue experimentation” to reproduce the disclosed results. A key aspect of the legal backdrop to the argument was the Federal’s test set forth in In re: Wands, 858 F.2d 731 (Fed. Cir. 1988) to determine whether undue experimentation is needed. The Wands test examines: (1) the quantity of experimentation necessary, (2) the amount of direction or guidance presented, (3) the presence or absence of working examples, (4) the nature of the invention, (5) the state of the prior art, (6) the relative skill of those in the art, (7) the predictability or unpredictability of the art, and (8) the breadth of the claims.

Patents-in-suit. At issue were certain claims of Amgen’s U.S. Patent No. 8,829,165 (“the ’165 patent”) and U.S. Patent No. 8,859,741 (“the ’741 patent”), both titled “Antigen binding proteins to proprotein convertase subtilisin kexin type 9 (PCSK9).” PCSK9 refers to a protein that impairs the body’s mechanisms for removing cholesterol. Amgen invented a class of monoclonal antibodies that bind to a “sweet spot” on PCSK9 and thereby inhibit PCSK9 from binding with LDL receptors. Amgen’s research team designed protocols, using super-immunized mice, to generate and select antibodies with the shape and chemical complementarity to bind PCSK9’s sweet spot. The ’165 and ’741 patents contained claims to monoclonal antibodies that bind one (or more) of the amino acids in PCSK9’s sweet spot. The patents disclose 26 example antibodies, characterized by amino-acid sequence. The patents also disclose the results of Amgen’s x-ray crystallography studies on two antibodies—21B12 and 31H4—providing an atomic-level picture of where those antibodies bind to PCSK9. These can be used as “anchors” to identify other antibodies that bind on the sweet spot. Antibody 21B12 was the basis for Amgen’s Repatha® drug, the first PCSK9 inhibitor approved worldwide to treat high LDL.

The patents further set forth a step-by-step “roadmap” for generating antibodies beyond the 26 examples. The patents disclosed that Amgen had isolated many more antibodies that bind the sweet spot, and through mice testing had arrived at 384 that block the LDL receptors “well.” The patents instruct skilled artisans to make either antibody 21B12 or 31H4 using the amino-acid sequences the patents provide. Next, they direct scientists to inject PCSK9 into mice to generate antibodies that bind to PCSK9. Third, they teach using one of the anchor antibodies from step one in competition assays to identify the antibodies from step two that bind to PCSK9’s sweet spot. Finally, they teach running Amgen’s optimized blocking test to confirm that the antibodies from step three block PCSK9’s interaction with LDL receptors.

Enablement dispute. The primary issue in the case is whether these step-by-step instructions adequately enable skilled artisans to create the full range of the claimed genus of antibodies without having to engage in “undue experimentation.” The parties seemed to agree on the basic points of law established by Section 112(a) of the Patent Act and relevant Supreme Court and Federal Circuit precedent. Along with factual differences (and disagreement as to the correct outcome), the parties tried to convince the Court that it could clarify the enablement standard and thereby provide guidance to lower courts and to innovators in the sciences. In particular, they disagreed as to what level of experimentation is “undue,” as well as the specific role of a “reasonableness” factor in determining what artisans would consider “important” enough to engage in experimentation.

Petitioner’s argument. Jeffrey Alan Lamken (MoloLamken LLP) kicked things off for the petitioner, Amgen. Lamken explained Amgen’s rationale for contending that the patents-in-suit were enabled. “The roadmap in Amgen's patents allows skilled artisans to easily make those antibodies every time using two new anchor antibodies that cover the entire sweet spot so skilled artisans can be certain to make all the claims' antibodies,” he said. “The Federal Circuit here never identified a single actual antibody that's in the claims that can't be made or requires undue experimentation. Instead, it invoked something that no one will defend is even relevant here: the cumulative effort to make all or some large group of an invention's potentially myriad variations.” Lamken argued that for a defendant to successfully assert an enablement defense, they must (1) first put forth evidence of a variant or category of the invention that requires what the Supreme Court has called “painstaking experimentation” and (2) and, second, if they meet the first prong, show why that matters to skilled artisans, which he characterized as a “reasonableness” component.

Justice Thomas opened questioning by asking Lamken to “tell us exactly what the invention is,” and to explain whether the 26 example antibodies were in fact all Amgen had invented. Lamken disagreed and pointed out that the patent stated that Amgen had identified 384, but only 26 were specified by amino acid structure as examples.

Justice Thomas asked whether the process could only create that limited number; Lamken said no, “it will produce every antibody within the claims.” Justice Thomas pressed on, asking, “So, in other words, you can't say how many?” Lamken disagreed and said, “the evidence shows in this art that about 400 you would get from -- coming out of the mouse.” Justice Thomas responded, “I think you’re making the point … in saying that, you don’t know how many there are.” Lamken replied that the patents indicated that the claimed process was routine—involving an easy method of swapping one amino acid for another—so the difficulty of making any next antibody was straightforward.

Justice Sotomayor queried, “If it’s so easy, why haven’t you made the 400 [sic]?” She also asked, “And did it take you so long to do the post-filing discovery of more?” Lamken said that a skilled artisan “isn't looking for every possible antibody. They're just looking for ones that bind to the right place and, therefore, block. And so, once you get those, your job is done.”

Justice Sotomayor pressed Lamken on the point that the patent, to her, “sounds to me like it’s all about just process,” and she pointed out that a chart supplied by Amgen didn’t indicate 26 “sweet spots.” Lamken explained Amgen’s position that the chart shows that they all can be made, supporting their contention of full enablement.

Justice Gorsuch then began a short line of questioning. First, Lamken agreed with Gorsuch that the law says “that a patent fails the enablement test if it would force a person skilled in the art to undertake undue experiment to produce the claimed invention,” although Lamken preferred to call it “painstaking experimentation.”

Justice Gorsuch next asked if Lamken considered the factors of the Wands test useful; Lamken said yes, when they are properly applied. “The problem with the Wands factors is they become something of a checklist that's abstracted and therefore replaces the ultimate statutory standard,” Lamken said. According to Lamken, the statutory standard is that of a skilled artisan, and what is required to “reasonably” enable them to make and use the invention.

In response to Justice Gorsuch’s next question—“[D]o you agree that the broader the patent, the more difficult it is to prove enablement?”—Lamken again gave a conditional answer, saying “Not necessarily. “You could have a relatively broad patent and you just have to have enablement commensurate with its scope.” Lamken cautioned that “if you have lots of categories within that patent, then you would have to enable what is important to the artisan within the category.”

Justice Gorsuch then questioned whether the broader the patent, the harder it was to prove enablement; that is, the more you have to show to indicate what skilled artisan has to do. Lamken said it was difficult to agree with that statement in the abstract because it depended upon “reasonableness” in the context of the specific art.

Chief Justice Roberts asked Lamken to flesh out the term “reasonably.” According to Lamken, it means what’s important to skilled artisans. “If you can find just some oddity that can't be made, that doesn't invalidate the patent because we're looking at what's important to skilled artisans.” He reiterated his view that the question is whether undue (or painstaking) experimentation was needed, and then whether that impeded the artisan from reasonably practicing the full scope of the invention. As far as how much experimentation was unreasonable—such as how much trial and error was required to reach the claimed results—Lamken said the answer is always “it depends,” explaining, “You're looking at the skilled artisan and you're saying what is a skilled artisan in this art willing to do.”

Raising an issue that seemed to resonate with some other Justices, Justice Kagan asked, “[D]o you understand the parties now all to agree on the appropriate legal test, and are we simply arguing now about how that test applies in this case?” Lamken allowed that the parties generally agreed on the applicable legal standards, and Justice Kagan followed up by asking, “[P]utting aside any application to this test, what do you think the parties don't agree on at this point with respect to principles of law?” Lamken answered, “I think the differences are as follows: The government would propose a requirement that you have a structure that unifies your genus, and I don't think that can be sustained under the law.”

Justice Gorsuch joined this line of querying and asked, if there aren’t other disagreements of law, “why isn’t this a fact bound dispute?” Lamken argued that the case wasn’t fact-bound because the Federal Circuit appeared to (wrongly) hold that “the evidence showed that the scope of the claims encompasses millions of candidates, and it would be necessary to first generate and then screen each candidate antibody to determine whether it meets the double function limitations,” and the Federal Circuit’s requirement that enablement would require structural disclosure for all of those millions, and “that can’t be right.” He urged the Court to bring clarity to the issue because the lower court’s interpretation regarding the “cumulative effort” to make all variations would be cost-prohibitive and would harm innovation.”

“The key is, on this,” said Lamken, “if there's uncertainty about what the Federal Circuit did or are doing, the answer is actually to bring clarity." The Court can provide guidance on the perspective of the skilled artisan, which involves a reasonableness standard. He repeated that a standard based on cumulative effort was incorrect and said the Court should make that clear.

Respondents’ argument. Paul D. Clement (Clement & Murphy, PLLC) argued on behalf of Sanofi and the other respondents. “Section 112 sets forth the heart of the patent bargain: The more you claim, the more you need to enable,” he began. “If you claim a lot and enable a little, the public is short-changed, and the patent is invalid. The Federal Circuit has long enforced that basic principle by requiring the patentee to enable the full scope of the patent without undue experimentation.” He noted that Amgen doesn’t take issue with that basic test or with the Wands factors, “or the vast bulk of the Federal Circuit's enablement precedent,” including that the full scope of the claimed invention must be enabled. According to Clement, “Amgen claims antibodies that bind on 16 residues in the epitope, but their specification does not enable skilled artisans to reliably produce them when they bind at 10 or more. They've overclaimed, they've underenabled, their patent is invalid.”

Justice Thomas asked Clement to restate or expand on what the claim is. Clement told Justice Thomas that the number 384 mentioned by Lamken was not the full scope of what Amgen claimed. “The numbers don't lie. They have claimed millions and millions of antibodies. And their reassurance that, don't worry, all of those millions that you get with conservative substitution, they're all going to work the same, that's inconsistent with their own expert's testimony in the Court below.”

Clement called special attention to an amicus brief by Sir Gregory Winter, a Nobel Prize-winning molecular biologist and renowned expert on the therapeutic use of monoclonal antibodies. According to Clement, Winter “will tell you that you can't look at function, and part of the problem here is these are purely functional claims. You can't look at function and say, oh, that tells me about the structure of antibodies that are going to bind and block in the right way, and you also can't look at the structure of one antibody and say, oh, if I just tweak it a little bit, it's going to do exactly the same thing.” And while cumulative effort is not itself a dispositive factor, it is appropriate to consider, Clement argued. In Clement’s opinion, Amgen’s supposed “roadmap” is mere trial and error.

Justice Gorsuch asked Clement if he agreed with Lamken that reasonableness was part of the standard, and Clement said yes, reasonableness was “just maybe the flip side of undue experimentation.” This promoted Justice Gorsuch to query, “So if we agree on the law, what’s left for this Court?” Clement quipped that maybe there wasn’t much except for a “DIG” (dismissal of the case as improvidently granted), which elicited laughter.

Justice Kagan also asked whether counsel disagreed on any other points of law, and Clement repeated his contention that the Amgen patents’ roadmap was inadequate. Justice Kagan asked whether that was really a disagreement about what the legal principles or standards are, and Clement said he thought so, opining that Amgen’s trial-and-error approach “consigns people skilled in the art to Sisyphean tasks forever.” He pointed again to the Winter brief, homing in on the assertion that the patents’ roadmap is not a shortcut at all but only “describes the routine processes that people use to make independent inventions.”

According to Clement, “there's no meaningful structure in these genus claims.” He elaborated by contending, “[W]hat you have before you is a particularly pernicious kind of claim. Because not only is it a genus claim that's purely functional or double functional, as the Federal Circuit described it, but it's really a work-around of [the Court’s holding in Association for Molecular Pathology v. Myriad Genetics, Inc., 569 U.S. 576 (2013), in which the Court determined that isolation of a naturally occurring segment of DNA was not patent-eligible]. Because basically they're pointing to something that exists in nature and they’re saying, we claim everything that works to bind there en bloc.”

In response to a question from Justice Jackson as to whether limiting the amount of acceptable experimentation would hinder some fields of study, Clement explained that the test should be “undue” and not a zero tolerance standard. Starting with the idea that the patent must enable a skilled artisan to make and use the invention, and if the patentee claims a lot, they have to enable of full scope of what is claimed.

Government’s views. Colleen R. Sinzdak, Assistant to the Solicitor General, presented the views of the U.S. government. It’s worth noting that while the Court was considering whether to grant certiorari, the government advised against it in an amicus brief. In that brief, the government rejected the proposition that the degree of experimentation required to implement the full scope of a patent’s claims is irrelevant to the enablement inquiry. “The Patent Act requires a patent to enable the ‘invention,’” the government said. “Thus, where a patentee purports to invent an entire genus, it must enable the entire genus.”

Before the Court, Sinzdak began by repeating a point made by Clement: “[I]f an antibody has already been created, a scientist who wants to make that antibody is not going to go into a laboratory and inoculate a mouse. They're going to use the amino acid sequence. That is the recipe for making an antibody. That is why the government says that, for the 26 exemplars within the patents, where Amgen had actually listed the amino acid sequence, those antibodies are enabled, because if a scientist wants to go into the lab and it wants to make that antibody, it has the recipe, it has the amino acid sequence.” Therefore, she contended, “they have not given the information that a person skilled in the art would need to make and use all of the antibodies within the genus.”

Justice Gorsuch again asked about the appropriate legal standard, specifically whether a reasonableness factor created some “elbow room” in the full-enablement requirement. Sinzdak said that in her view the need for tempering by “reasonableness” was taken care of by the doctrine of equivalents. Justice Gorsuch also asked about the issue of cumulative effort, and Sinzdak responded by emphasizing the very broad scope of the patent. She said that she was not sure that even 384 antibodies can be made via the mouse process, and the patents claimed antibodies made by any process. In her view, this kind of broad genus claims harmed innovation.

Justice Sotomayor reminded Sinzdak about Clement’s remark that the Court should “DIG” the case, adding, “If we didn't want to, what could we say to have the Federal Circuit or anyone else who's interested in this area?” Sinzdak said, “We do not think that the Federal Circuit said anything wrong here. I think that some of the arguments that we're hearing from Petitioners suggest that it might be useful to clarify that you really do need to enable each of the different embodiments.” However, she rejected Lamken’s contention that one could disregard portions of the claimed scope because they “don’t matter” to artisans. “And, of course, what they're saying is it doesn't matter, but, by the way, if you do find it and it does something truly amazing, we own it.”

Justice Kavanaugh suggested that it might be helpful for the Court to say that it agreed with the Federal Circuit’s approach and to make it clear that if billions of research dollars were at stake because of this approach, it was up to Congress to change the law. Sinzdak agreed, also pointing out that if Amgen were asking the Court to create a special exception to the enablement rules for antibodies, that was something for Congress, not the Court.

The case is No. 21-757.

Attorneys: Jeffrey Alan Lamken (MoloLamken LLP) for Amgen Inc. Paul D. Clement (Clement & Murphy, PLLC) for Sanofi, et al. Colleen R. Sinzdak, U.S. Department of Justice, for the United States.

Companies: Amgen Inc.; Amgen Manufacturing, Ltd.; Amgen USA, Inc.; Sanofi, Aventisub LLC, f/k/a Aventis Pharmaceuticals Inc.; Regeneron Pharmaceuticals Inc.; Sanofi-Aventis U.S. LLC

MainStory: TopStory Patent GCNNews